New Study: Do Fruit Flies Have Free Will?

MSNBC is reporting on a new study that indicates a potential basis for free will in fruit flies. Philosophers break down into a trilogy of attitudes about free will. On the one hand there are determinists who believe that universe is essentially governed by constant rules. Incompatibilist determinists believe that as a result there can be no such thing as free will, while compatibilists believe that there can be free will even in a deterministic universe. Then there are those who believe the universe, or at least human beings, aren't deterministic at all. Neurobiologist Björn Brembs gives a succinct summary of the fundamental contradiction between the major views of incompatibilist determinism and non-determinism:

Free will is essentially an oxymoron — we would not consider it 'will' if it were completely random and we would not consider it 'free' if it were entirely determined.

For this reason the compatibilist position is the one that enjoys the most widespread support from lay people even though it is actually perhaps the hardest to defend. Noted philosopher of the mind Daniel Dennett is a major proponents of this theory in books such as "Elbow Room". Dennett argues, as the name dictates, that there is a certain type of limited freedom to be found within a determinism that is sufficient to meet our intuitive desires to think of ourselves as "free". Interestingly enough Daniel Dennett uses the seemingly robotic behavior of the wasp Sphex as an example of the kind of determinism that humans want to avoid, and this most recent study returns to the insect world with precisely the opposite intent: to locate some spark of free will in the behavior of some of the simplest creatures (in this case the fruit fly).

The overall intent of the study is generally the same as Dennett's, however, to find some sort of wiggle room for free will within determinism, as described by mathematical biologist George Sugihara "We speculate that if free will exists, it is in the middle ground."

In order to test this hypothesis, an experiment was conducted to monitor the behavior of fruit flies in completely plain white rooms that offered no motivation to move in any particular generation. Bereft of any meaningful input, the hypothesis was that any differences between the behavior of one fruit fly and the next would be completely random if the flies had no free will. This is because any differences would be chalked up to errors within the brain of the fruit fly.

What the witnessed, however, was that the behavior of the fruit flies was "far from random". Sugihara says there seems to be "a function in the fly brain which evolved to generate spontaneous variations in the behavior." This, however, is where the MSNBC article seems to break down. The behavior appears to match Levy's distribution. Not only is Levy's distribution not an algorithm, as the article reports, but it is a distribution of a random variable. It is, in short, random. Most likely when the article refers to random it specifically refers to the continuous uniform distribution. The difference between a uniform distribution and another type of distribution is easy to visualize. If you roll one regular die, you will get a number between 1 and 6 and each number is equally likely. That's a uniform distribution. If you roll two regular die and add the sum you will get a number between 2 and 12, but the numbers will not be equally likely. A 2 and a 12 are the least frequent while a 7 is the most common. Which number you get is still random, but now there is a pattern to the randomness.

So it's not clear what the scientists have proved by demonstrating that the fruit flies' random behavior matches Levy's distribution and not a uniform distribution. It is, after all, still random. The behavior arises spontaneously and, according to the article in the journal PLoS ONE, this means the behavior is something between completely random and purely determined. Says Sugihara, this "could form the biological foundation for what we experience as free will."

But it seems that MSNBC is a little over-eager to report these findings as a breakthrough. Neuorscientist Gonzalo de Polavieja specifies that what we've really found is "a complex decision-making processing underlying behavior. This seems a necessary condition for free will." Or, of course, just a really smart robot. Brembs goes even further by stating that the study "only showed that brains might possess a faculty which free will could potentially be based on".

I'm skeptical that this study has any significant impact on the free will argument whatsoever. Even if the behavior of the flies differed in non-random ways from fly to fly, there's no way to be sure this isn't merely expressing physical differences in the brains of the respective flies. More fundamentally, however, the behavior is random and so it doesn't seem to have any significant impact on the compatibilism vs. incompatibilism debate. Like Dennett's elbow room, scientific efforts to resolve the free will debate seem to bring us, in the words of Professor James Hall, "pretty thin soup indeed."